{
  "id": 464398,
  "title": "Are we allowed to do inverse engineering?",
  "url": "/competitions/2023-gwdata-bootcamp/discussion/464398",
  "author_name": "",
  "post_date": "2023-12-30T09:05:45.924552900Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi,</p>\n<p>As the data is generated, can we just use the math behind and solve the optimizer given our objective function directly? Or training a ML model is a must to solve this challenge?</p>\n<p>Thanks</p>",
  "messages": [
    {
      "id": "2579708",
      "postDate": "12/30/2023 09:05:45",
      "content": "<p>Hi,</p>\n<p>As the data is generated, can we just use the math behind and solve the optimizer given our objective function directly? Or training a ML model is a must to solve this challenge?</p>\n<p>Thanks</p>",
      "rawMarkdown": "Hi,\n\nAs the data is generated, can we just use the math behind and solve the optimizer given our objective function directly? Or training a ML model is a must to solve this challenge?\n\nThanks",
      "votes": null
    },
    {
      "id": "2582543",
      "postDate": "01/01/2024 16:44:50",
      "content": "<p>Hi, <a href=\"https://www.kaggle.com/starplatinum810\" target=\"_blank\">@starplatinum810</a> </p>\n<p>Absolutely, I believe that's a feasible approach. We are open to any method that yields effective results, so you're not restricted to using machine learning models if you find an alternative solution more suitable. It's all about the outcome! Just a reminder, after the competition concludes, please ensure to share your reproducible code. Are you considering using some GW search pipeline for this challenge? That sounds like an interesting strategy!</p>\n<p>Best regards,</p>\n<p>He Wang</p>",
      "rawMarkdown": "Hi, @starplatinum810 \n\nAbsolutely, I believe that's a feasible approach. We are open to any method that yields effective results, so you're not restricted to using machine learning models if you find an alternative solution more suitable. It's all about the outcome! Just a reminder, after the competition concludes, please ensure to share your reproducible code. Are you considering using some GW search pipeline for this challenge? That sounds like an interesting strategy!\n\nBest regards,\n\nHe Wang",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2582543,
      "author_name": "herbwang",
      "author_url": "",
      "post_date": "01/01/2024 16:44:50",
      "content": "<p>Hi, <a href=\"https://www.kaggle.com/starplatinum810\" target=\"_blank\">@starplatinum810</a> </p>\n<p>Absolutely, I believe that's a feasible approach. We are open to any method that yields effective results, so you're not restricted to using machine learning models if you find an alternative solution more suitable. It's all about the outcome! Just a reminder, after the competition concludes, please ensure to share your reproducible code. Are you considering using some GW search pipeline for this challenge? That sounds like an interesting strategy!</p>\n<p>Best regards,</p>\n<p>He Wang</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2579708": "Hi,\n\nAs the data is generated, can we just use the math behind and solve the optimizer given our objective function directly? Or training a ML model is a must to solve this challenge?\n\nThanks",
    "2582543": "Hi, @starplatinum810 \n\nAbsolutely, I believe that's a feasible approach. We are open to any method that yields effective results, so you're not restricted to using machine learning models if you find an alternative solution more suitable. It's all about the outcome! Just a reminder, after the competition concludes, please ensure to share your reproducible code. Are you considering using some GW search pipeline for this challenge? That sounds like an interesting strategy!\n\nBest regards,\n\nHe Wang"
  },
  "source": "meta"
}